The progression of DCIS to invasive carcinoma is a poorly understood key step in breast tumor progression. We have previously demonstrated that molecular changes occur in all cell types during tumor progression and that the loss of myoepithelial cells leads to invasive tumors in MCF10DCIS xenografts. We have characterized the comprehensive gene expression profiles of putative progenitor-like and differentiated myoepithelial cells from normal human breast tissue and identified markers that can be used for their identification. Based on our data we hypothesize that the differentiation of progenitors to myoepithelial cells is progressively inhibited by signals coming from tumor epithelial cells and cells composing the microenvironment. As a consequence of this, the myoepithelial cell layer is gradually lost and eventually disappears, leading to transition to Invasion. We propose three specific aims to test these hypotheses. Aim 1: To identify molecular determinants of normal myoepithelial cell differentiation and their abnormalities in DCIS. We will isolate CD10+ subpopulations of cells from normal breast tissue of healthy women and BRCA1 mutation carriers, and from different subtypes of DCIS. Molecular profiles of each of the cell populations will be characterized using genome-wide approaches followed by validation in a larger set of samples at the single cell level. Aim 2: To determine if abnormalities of myoepithelial differentiation in DCIS correlate with the risk of progression to invasive disease. We will determine if the expression of regulators of myoepithelial cell differentiation in DCIS is associated with invasive progression. Aim 3: To test if modulating signaling pathways involved.in myoepithelial cell differentiation would effect progression to invasion. Using primary cell culture models, we will determine if perturbations of regulators of myoepithelial cell differentiation lead to loss of myoepithelial cells and progression to invasion. Identification of determinants of normal myoepithelial cell differentiation and their abnormalities in DCIS would improve our understanding of stem cell differentiation and provide novel prognostic markers and targets for therapeutic and preventative interventions.